9. Misspecification and data issues

Question 1

When MLR.4 holds, what do we say about the explanatory variables?

Answer 1

When MLR.4 holds we sat that we have exogenous explanatory variables

Question 2

When do we call our explanatory variables endogenous?

Answer 2

If xj is correlated with u for any reason then xj is said to be endogenous explanatory variable

Question 3

Why is MLR.4 (Zero conditional mean) not holding important?

Answer 3

Because we no longer have unbiased estimators

Question 4

What are the three reasons why an explanatory variable could be exogenous?

Answer 4

• Omitting a key variable
• Functional form misspecification
• Measurement error

Question 5

What is excluding a relevant variable called?

Answer 5

Underspecification of the model

Question 6

What is the impact of overspecifying the model?

Answer 6

It lowers precision

Question 7

What does under specifying the model cause?

Answer 7

It will cause all the OLS estimators to be biased an inconsistent

Question 8

If we assume that two of our explanatory variables may be correlated what do we assume about them?

Answer 8

If x1 and x2 are correlated, assume a linear regression relationship between them

Question 9

What is functional form misspecificiation?

Answer 9

When a model does not properly account for the relationship between the dependent and the observed explanatory variables

Question 10

How can we test for functional form misspecification?

Answer 10

Using F-tests for joint exclusion restrictions (L4)

Question 11

How does an F-test for joint exclusion restrictions work?

Answer 11

• Add quadratic terms of any significant variables to a model
• Perform a joint test of significance on all quadratic terms
• If they are significant they can be added to the model

Question 12

What is the full name of a RESET test?

Answer 12

Regression specification error test

Question 13

What is the idea of a RESET test?

Answer 13

The idea is to include squares and possibly higher order fitted values in the regression

Question 14

What does a rejection of the null hypothesis under the RESET test tell us?

Answer 14

That there is evidence of joint significance and therefore evidence of misspecification

Question 15

What should we do if we reject the null in our model?

Answer 15

Try something else, log them for example

Question 16

What does the Davidson-MacKinnon test do?

Answer 16

Tests whether you should take the log form or not

Question 17

What are lagged variables?

Answer 17

The same value as your dependent variable but from an earlier time period

Question 18

Define measurement error

Answer 18

Difference between the observed value y and the actual value y*

Question 19

What is an indicator variable?

Answer 19

A binary variable

Question 20

What should you do if you are missing data?

Answer 20

Use the missing indicator method

Question 21

What should you do if you have large outliers in your model but you don’t want to take them out?

Answer 21

Use the least absolute deviation estimation

Question 22

Why can large outliers cause problems for OLS?

Answer 22

Because OLS works by taking the deviation from the average and squaring it which skews the results of OLS

Question 23

If you are struggling to find a proxy variable, what could you use?

Answer 23

A general possibility is to use the lag of the dependent variable

Question 24

How can we try and solve the omitted variables problem?

Answer 24

Using a proxy variable

Question 25

What are the advantages of a RESET test?

Answer 25

It is easy to apply because it does not require one to specify what the alternative model is

Question 26

What are the disadvantages of a RESET test?

Answer 26

We only know whether the model is misspecified or not, but it does not help us necessarily in choosing a better alternative